UK Court Sides Mostly with Stability AI in Getty Case

Ruling narrows legal questions around AI training and copyright

A British High Court has largely ruled in favor of London-based artificial intelligence firm Stability AI in a closely watched legal dispute with Getty Images. The case centered on allegations that Stability used 12 million Getty-owned images without permission to train its image generation model, Stable Diffusion. While Getty secured a partial win on trademark grounds, the court dismissed the bulk of its copyright claims. Both companies issued statements framing the outcome as a victory.

Copyright Claims Dismissed, Trademark Partially Upheld

Getty had accused Stability of both primary and secondary copyright infringement, arguing that the company unlawfully used its image library to train an AI model. Primary infringement involves direct reproduction, while secondary refers to distributing or importing infringing content. During the trial, Getty dropped its primary claim, acknowledging it was unlikely to succeed, but continued to argue that offering Stable Diffusion in the UK amounted to importing unauthorized copies. Justice Joanna Smith rejected this, ruling that the AI model does not store or reproduce copyrighted works.

The court did find limited instances of trademark infringement, noting that Getty’s watermark appeared on some AI-generated images. However, the judge emphasized that these occurrences were not widespread and described the ruling as “historic and extremely limited in scope.” Stability welcomed the decision, stating that it resolved the core copyright concerns. Getty, while disappointed by the broader outcome, highlighted the trademark ruling as a win for intellectual property owners.

Broader Legal Uncertainty Remains

Legal experts say the ruling leaves unresolved questions about the legality of training AI models on copyrighted material. Iain Connor, an intellectual property lawyer, noted that the UK still lacks a definitive judgment on whether such training constitutes infringement. Justice Smith acknowledged the broader societal importance of the issue but said the court could only rule on the narrowed case presented. The decision underscores the difficulty of applying existing copyright frameworks to emerging AI technologies.

Getty has not confirmed whether it will appeal the UK ruling. Meanwhile, the company is pursuing a separate copyright lawsuit against Stability in the United States, which was refiled in San Francisco in August. That case is part of a growing wave of litigation involving generative AI and the use of creative works in training datasets. Courts on both sides of the Atlantic are now grappling with how to balance innovation with the rights of content creators.

Industry-Wide Legal Pressure Mounts

Stability’s case is one of more than 50 ongoing copyright lawsuits targeting AI companies. The legal landscape is evolving rapidly, with some cases resulting in settlements and others being dismissed. Anthropic recently agreed to a $1.5 billion settlement with a group of authors, while a similar case against Meta was thrown out. Major studios including Warner Bros., Disney, and Universal have also filed suits against image generator Midjourney, alleging unauthorized use of copyrighted characters.